Overview and details of the sessions of this conference. Please select a date or room to show only sessions at that day or location. Please select a single session for detailed view (with abstracts and downloads if available).
5h. AMR: Determinants and Surveillance of Antibiotic Use in the Community
Predictors of Antibiotic Use in African Communities: Evidence from Household Surveys in Five African Countries
1Harvard Medical School, United States of America; 2World Health Organization, Regional Office for Africa, Brazzaville, Congo
Problem statement: Evidence about antibiotic overuse is lacking in low-income countries where reliable antibiotic consumption data are not available.
Objective: To generate reliable evidence on antibiotics use at the consumer level. Specific aims are to describe how antibiotics are used in African households and to identify key predictors of antibiotic use.
Design: Descriptive, cross-sectional analysis of survey data
Setting: Household surveys conducted in five African countries between 2007 and 2008 with an instrument developed by the World Health Organization (WHO) to monitor country pharmaceutical situations at the community level.
Study population: Households were selected by multistage cluster sampling (900 to 1,080 households per country). Study population consisted of 2,914 household members with an acute illness in the two weeks preceding the survey; data were collected on all sick members in 3 surveys (Ghana, Kenya, Uganda) or only on the youngest sick member in two surveys (Gambia, Nigeria).
Outcome measures: Care seeking outside home for acute illnesses, use of antibiotics to treat acute illnesses
Results: Half of the households reported at least one recent acute illness; 19% of illnesses were considered very severe and 48% somewhat severe. A large proportion of sick individuals (90%) sought care outside home, an even greater proportion (95%) took medicines, and 36% took antibiotics. Patients were more likely to seek care outside home if they felt that the closest public facility usually had medicines or that they could usually afford medicines. In adjusted multivariate analyses, the strongest predictor of antibiotic use was the presence of upper respiratory (UR) symptoms (OR: 3.02, CI: 2.36–3.86). Patients who sought care outside home were significantly more likely to receive antibiotics if they had diarrhea or difficulty breathing, but significantly less likely if they had fever. The perception of a severe illness was a strong predictor of seeking care outside home (OR: 3.24, CI: 1.69–6.22). For those seeking care, however, the likelihood of receiving antibiotics was independent of illness severity and highest for patients visiting public hospitals.
Conclusions: Our results provide direct evidence about care seeking for acute illness and about community consumption of antibiotics in African countries. They underscore the degree to which antibiotics are misused by consumers, especially for UR symptoms. They highlight the need to educate prescribers, dispensers, and consumers since antibiotics are widely and inappropriately used in all settings.
Funding source(s): The WHO Department of Essential Medicines in Geneva organized and funded data collection, with support from the Medicines Transparency Alliance. The WHO African Regional Office funded the study.
Surveillance of Antimicrobial Use in Resource-Constrained Community Settings
1World Health Organization, Regional Office S.E.Asia, India; 2World Health Organization, Geneva; 3Nelson R Mandela School of Medicine, University Kwa Zulu Natal, S Africa
Problem statement: In resource-constrained settings, antibacterial medicines (ABM) are frequently inappropriately used, contributing to increased antimicrobial resistance. There are few sources of good data on ABM use in such settings, nor are there standard methods for conducting such surveillance to inform public health decision making.
Objectives: To assess the feasibility of surveillance of ABM use in the community in 3 resource-constrained settings in India and 2 in South Africa
Design: Time series data on monthly ABM use were collected over at least 12 months at each site. Each site sought to document 30 patient encounters where ABMs were provided from 7–30 facilities per month.
Setting: Public and private sector primary care facilities in 3 sites, public facilities in one site, and private pharmacies in another site
Study population: ABM use data were collected monthly at the 3 Indian sites by interviewing patients exiting from facilities and from the 2 South African sites by extracting data from patient records in the facilities. In addition, 2 Indian sites collected procurement and sales data from public facility records and by interviewing private retailers.
Outcome measure(s): (1) Percentage patients receiving a specific ABM, (2) number of defined daily doses (DDD) of a specific ABM per 100 patients attending the facility per month, and (3) lessons on setting up surveillance
Results: There was extensive use of ABMs in all sites. Older agents such as co-trimoxazole were used more in public facilities and newer agents such as fluoroquinolones in private facilities. Although methodological differences limit comparability of data, use appeared to be higher in India as compared to South Africa for all facility types. It was easier and more reliable to measure ABM use as the percentage of patients receiving an ABM than to determine DDDs per 100 patients per month, although the latter gave more information about dosage and duration. Conducting patient exit interviews was more resource intensive than extracting data from patient records, but the latter was dependent on the completeness of records and could not capture over-the-counter sales. The 2 sites that collected sales and purchase data felt that it was not as reliable as collecting individual patient data. All sites complained that facilities, particularly in the private sector, became fatigued by the data collection process.
Conclusions: All 5 pilot sites provided useful data on ABM use but also raised a number of technical and logistical issues related to long-term surveillance in resource-constrained community settings. ABM use measured as the percentage of prescriptions containing a specific ABM is easier and more reliable to use in these settings than DDD methodology.
Funding source(s): WHO, USAID. Study reported on behalf of the Surveillance of Antimicrobial Use and Resistance in Resource-Constrained Settings Project Group.
Behavioural Model for Community-Based Antimicrobial Resistance, Vellore, India
1World Health Organization, Switzerland; 2Nelson Mandela School of Medicine; 3Christian Medical College Hospital Vellore
Problem statement: Resistance to antimicrobial agents compounds the burden of diseases worldwide. Difficulties to estimating the impact of AMR on individuals and the community or the impact of AM use on resistance in resource-constrained settings is compounded by the paucity of community-based data. Robust surveillance data collection methodologies are lacking in such settings. More explorations and improved analytical methods are needed to fully understand trends and impact of AMR on cost of illness and to inform AMR surveillance.
Objectives: To determine the behavioural trends—seasonality (periodicity) and the temporal associations—between community-based AMR and AM use; to forecast the short-run pattern in AMR through the behaviour of AMR and the predictors (indicators) of AMR; and to compare the temporal correlation of the trends in DDD and the proportion of patients prescribed antibiotics, with community based AMR
Design: Longitudinal, non-comparison time-series
Settings: A multi-centre WHO study in India and South Africa. We use AMR surveillance data obtained from Vellore (urban area) and Kuppam (rural area) with a combined population of 500,000 within Vellore District, Tamil Nadu State of southern India.
Study population: Study isolated commensal E. coli (N=2,026) from pregnant women attending antenatal clinics. Monthly AM-use data were obtained from exit interviews from hospitals or PHC clinics (including not-for-profit and for-profit hospitals in the urban area and public sector PHC clinics and a not-for-profit hospital in the rural areas); private sector pharmacies; and private sector general medical practitioners’ practices. Prescriptions containing antibiotics totaling 21,600 were obtained from 52,788 prescriptions. Data were collected in two time periods, from August 2003 to July 2004 and from January to December 2005.
Outcome measure(s): Monthly proportion of isolated E. coli resistant to co-trimoxazole, extended spectrum penicillin, and quinolones; monthly antibiotic use in DDD and proportion of patients prescribed antibiotics
Results: Both AMR and AM use demonstrated lagged trends and seasonality. AMR lags vary between 3 and 5 months of AM use. Impulse-response could last as much as 15 to 45 months. AM use demonstrated significant Granger-causality with AMR in addition to circularity. Both monthly DDD per patient and proportion of patients on specific antibiotics show similar effects on AMR, but DDD per patient appear to demonstrate more reactive effect on AMR.
Conclusions: Community AM use can predict AMR. Our results provide additional evidence for estimating the economic impact of AMR and could inform the design of community-based antimicrobial surveillance and interventions in low-resource settings.
Funding source(s): WHO
Identifying Key Beliefs of Self Medication with Antibiotics in Yogyakarta City, Indonesia
1Faculty of Pharmacy Sanata Dharma University Yogyakarta, Indonesia; 2Faculty of Medicine Gadjah Mada University Yogyakarta, Indonesia; 3School of Nursing, University of Adelaide Australia; 4Discipline of Public Health, University of Adelaide Australia; 5Faculty of Health Sciences, Australian Catholic University Australia
Problem statement: Although antibiotics in Indonesia are categorized as prescription-only medicine, self medication with antibiotics (SMA) is common. Beliefs about SMA that might influence SMA behaviour remain unexplored, however.
Objectives: To identify key beliefs of SMA based on theory of planned behaviour (TPB), including behavioural, normative, and control beliefs and associations between these beliefs and intention of SMA
Design: This descriptive study used an interview guideline informed by TPB and a pre-tested questionnaire developed from findings of the preceding interviews.
Setting: This study was a population-based study.
Study population: Population included adults (>18 years old). Snowball and cluster random sampling were applied to select 25 participants for semi-structured interview and 640 participants for self-administered questionnaire, respectively.
Outcome measure(s): Key beliefs of SMA and their associations with intention of SMA
Results: In total, 25 face to face interviews have been conducted. Participants reported that advantages of using non-prescribed antibiotics were to save money and time—as a result of avoiding a medical consultation—and to avoid taking too many types of medicines commonly prescribed by doctors. Fear for adverse effects, poor outcome, and antimicrobial resistance were declared as disadvantages. Availability of antibiotics to be purchased without prescription in pharmacies, drug stores, and shops/kiosks, and previous successful medication use made this behaviour easier. Participants tended to seek advice from medical practitioners for their children’s health concerns, however. Family members and friends, especially those with a health education background, were more likely to approve of this behavior. Results of self-administered questionnaire (n = 283 participants who were familiar with antibiotics) showed that, as expected, a range of beliefs of SMA significantly correlates with intention to do SMA. Most participants were aware of the risks of SMA, had no pressure from their social networks to practice SMA, and were reluctant to obtain non-prescribed antibiotics from shops/kiosks. Participants would more likely to have intention to do SMA when they have previous successful experience in using antibiotics and they can purchase antibiotics without prescription from pharmacies with odds ratios of 0.27 (0.06–0.95) and 0.15 (0.03–0.81), respectively.
Conclusions: Variety of people’s beliefs in using non-prescribed antibiotics revealed by interviews was useful to generate a valid and reliable tool for investigating key beliefs of such use through population-based survey based on TPB. Findings suggest that strengthening people’s awareness regarding the harmfulness of SMA may be effective to discourage intention to do so. Since shops/kiosks are not popular as sources of non-prescribed antibiotics, efforts for improving SMA should be focused on pharmacies and drug stores. Underlying factors of such behaviour should be further investigated.
Funding source(s): The Indonesia Ministry of Education